基于随机森林算法的车辆行人保护头部损伤研究

在道路交通事故中,行人和骑行者的伤亡率较高,头部损伤是造成行人死亡的关键因素之一,提高行人头部保护能力是车辆行人保护技术发展的关键方向。当前,各大主机厂主要依赖CAE仿真技术开发行人保护能力,利用试验数据进行验证。本文将基于随机森林算法,开发出基于试验数据的行人保护结果预测模型,模型预测的准确度达80%。该预测模型能协助主机厂从试验数据的角度提高车辆行人保护性能的开发效率。

In road traffic accidents,the casualty rate of pedestrians and cyclists is high,and pedestrian head injury is one of the key factors causing pedestrian death,and improving pedestrian head protection capability is a key direction for the development of vehicle pedestrian protection technology.Currently,major OEMs mainly rely on CAE simulation technology to positively develop pedestrian protection capability and use test data for validation.In this paper,a pedestrian protection outcome prediction model based on test data is developed based on random forest algorithm,and the accuracy of the model prediction reaches 80%.The prediction model can assist OEMs to improve the development efficiency of vehicle pedestrian protection performance from the perspective of test data.